Source-grid-load-storage resource collaborative planning method oriented to supply-demand collaborative capability optimization

Abstract

This invention relates to the field of power system technology and discloses a resource coordination planning method for power generation, grid, load, and storage to optimize supply and demand coordination capabilities. The method includes: constructing a coordination capability evaluation index system to quantitatively assess supply and demand coordination capabilities; constructing a comprehensive optimization planning model based on the evaluation index system that considers both economic efficiency and supply and demand coordination capabilities; using minimizing the total life-cycle cost and maximizing supply and demand coordination capabilities as the comprehensive objective function; and solving the comprehensive optimization planning model to obtain the optimized solution. The advantages of this invention are that by strengthening the coordinated allocation of various resources, it improves the dynamic adaptability of system operation, and it can be extended to city-level and other power generation, grid, load, and storage related planning scenarios, providing methodological support for the clean and low-carbon transformation of energy systems.

Description

Technical Field

[0001] This invention relates to the field of power system technology, and in particular to a source-grid-load-storage resource collaborative planning method for optimizing supply and demand coordination capabilities. Background Technology

[0002] Driven by global energy transition goals, the integration of a high proportion of renewable energy into the power system has become an inevitable trend. Its core objective is to increase the share of clean energy and achieve carbon reduction in the energy system through end-to-end low-carbon synergy. However, the intermittent and fluctuating nature of renewable energy output contradicts the system's demands for reliability, economy, and low carbon emissions. This can easily lead to supply-demand mismatches, wind and solar curtailment, and peak-shaving pressures. Furthermore, the traditional "source follows load" dispatching model is ill-suited to the new characteristic of "load follows source," hindering the achievement of low-carbon goals and the safe and economical operation of the system. Against this backdrop, achieving a coordinated balance between power sources, energy storage, and load through scientific capacity optimization has become a core issue in the planning and operation of new power systems.

[0003] In view of the multi-timescale characteristics of power system supply and demand, existing single-layer, double-layer and multi-stage optimization models have decomposed planning and operation problems to achieve refined decision-making for supply and demand coordination. Reference [1] proposed a supply and demand coordination optimization model through the interactive and complementary relationship between electricity and heat energy on both sides of supply and demand; Reference [2] established a hierarchical optimization scheduling model to improve the flexibility of pumped storage system and achieved source-load coordination optimization through interactive iteration; Although such studies have achieved coordination between planning and operation through different architectures, they mostly focus on the optimization of single equipment or local benefits and lack deep coupling between system-level goals and coordination mechanisms.

[0004] In view of the dynamic characteristics of supply and demand coordination, operation simulation, as the core means of analyzing source-load interaction and resource regulation capacity, has become an important link in the verification of planning schemes. Reference [3] considers cross-border power grid interconnection and energy storage at different time scales, and establishes a source-grid-storage integrated planning model based on the annual time-series operation simulation; Reference [4] proposes a coordinated optimization configuration method based on a two-stage model of planning and operation simulation; Existing methods mainly adopt two approaches: based on typical daily scenarios or based on 8760h operation simulation. Although the typical daily scenario sampling method reduces the amount of computation by using clustering or hierarchical sampling, it is difficult to fully cover edge scenarios such as sudden changes in new energy output and extreme load values; Although the 8760h full-time simulation method retains complete time-series information, it suffers from problems such as low solution efficiency, long computation time, and difficulty in convergence due to high-dimensional data.

[0005] References

[0006] [1] Cheng Shan, Wei Zhaobin, Huang Tianli, et al. Optimized operation of microgrid based on multi-energy complementarity cogeneration [J]. Power System Protection and Control, 2020, 48(11):160-168. DOI:10.19783 / j.cnki.pspc.190932.

[0007] [2] Lü Wanyu, Zhao Hongsheng, Han Yingsheng, et al. Research on hierarchical optimization scheduling method to improve the flexibility of pumped storage system under supply and demand uncertainty [J / OL]. Journal of Electrical Engineering, 1-16 [2025-08-17].

[0008] [3] Huang Xuxiang, Han Xueshan, Li Jiawei, et al. Coordinated planning of energy storage and various power sources in large power grids [J]. Distributed Energy, 2019, 4(05):67-74. DOI:10.16513 / j.2096-2185.DE.191079.

[0009] [4]QI YC, HU W, DONG Y, et al. Optimal configuration ofconcentrating solar power in multienergy power systems with an improvedvariational autoencoder[J]. Applied Energy, 2020, 274: 115124. Summary of the Invention

[0010] The purpose of this invention is to overcome the shortcomings of the prior art and provide a source-grid-load-storage resource collaborative planning method for optimizing supply and demand coordination capabilities.

[0011] The objective of this invention is achieved through the following technical solution: a source-grid-load-storage resource collaborative planning method for optimizing supply and demand coordination capabilities, the method comprising,

[0012] Construct a collaborative capability evaluation index system to quantitatively assess supply and demand collaborative capabilities;

[0013] A comprehensive optimization planning model that balances economic efficiency and supply-demand synergy is constructed based on a collaborative capability evaluation index system; the comprehensive objective function is to minimize the total life cycle cost and maximize the supply-demand synergy.

[0014] The optimal solution is obtained by solving the comprehensive optimization planning model.

[0015] Specifically, the collaborative capability evaluation index system includes the following indicators: proportion of new energy installed capacity, proportion of new energy power generation, adequacy of installed capacity, adequacy of reserve capacity, power utilization rate, new energy utilization rate, cost proportion, proportion of adjustable power installed capacity, average adjustable capacity of adjustable power, adequacy of adjustable power capacity, proportion of imported power, average annual load satisfaction rate, energy storage configuration ratio, and average energy storage utilization rate.

[0016] Specifically, the indicators are weighted using a combined weighting method of CRITIC objective weighting and G1 subjective weighting. The CRITIC objective weighting method is used to obtain objective weights by accurately mining objective information, while the G1 subjective weighting method is used to obtain subjective weights by reasonably ranking the importance of indicators based on the decision-maker's experience.

[0017] Specifically, a linear weighting method is used, and the combined weight is obtained based on objective weights and subjective weights. The calculation formula is as follows:

[0018] ;

[0019] In the formula, For subjective weight vectors, ;w k For objective weight vectors, ; These are the combination coefficients; This is an unnormalized combined weight vector; represents the standardized combined weights; N is the length of the indicator matrix, and n is the indicator number.

[0020] Specifically, the objective function is:

[0021] ;

[0022] In the formula, For investment costs; Operating costs; For carbon emission costs; C buy For electricity purchase cost; C cut Cost of power curtailment; This is the upper limit of the cost. A quantitative value for supply and demand coordination capability; This is the comprehensive weight matrix of the indicators; This is a vector of index values; , The weights are respectively based on economic efficiency and the ability to coordinate supply and demand.

[0023] Specifically, the constraints of the comprehensive optimization planning model include:

[0024] System equilibrium constraints:

[0025] ;

[0026] In the formula, P buy,t The amount of electricity generated for which electricity is purchased; , , , These are, respectively, energy storage discharge power, pumped hydro storage discharge power, energy storage charging power, and pumped hydro storage charging power; This is the sum of the power generation of all power supply devices during time period t; The load shedding amount at each time point; The load power at each time point;

[0027] Shear load constraint:

[0028] ;

[0029] In the formula, This represents the maximum load shedding rate at each time point; This represents the overall maximum load shedding rate.

[0030] Conventional power supply dynamic constraints:

[0031] ;

[0032] In the formula, A collection of conventional power supply units; Let t be the online capacity of a certain power unit; , To enable or disable capacity; This refers to the rated capacity of a power unit. To provide power to a certain type of power unit; This represents the minimum technical output rate of a certain power unit. and These are the minimum and maximum ramp rates for this power unit;

[0033] New energy constraints:

[0034] ;

[0035] In the formula, For wind power output during period t; Provide photovoltaic power output for period t; , This represents the predicted power output value for wind and solar power during time period t.

[0036] Constraints of energy storage systems:

[0037] ;

[0038] In the formula, It combines electrical energy storage and pumped hydro storage; , These are the maximum power and capacity of the energy storage, respectively. Indicates the lower limit of energy storage capacity; , These represent the states of charge of a certain energy storage system at time t and the previous time, respectively. For time step; , The charging power at time t and time t-1 are respectively. , These represent the discharge power at time t and the previous time, respectively. Let be the state variables of the operating mode of a certain energy storage system during time period t; The overall energy conversion efficiency of a certain energy storage system.

[0039] Specifically, the nonlinear part of the comprehensive optimization programming model is transformed into a linear problem through linearization, and the calculation is as follows:

[0040] ;

[0041] In the formula, Let x be any real number; x is the decision variable; For the decision variable space; function , These are the numerator and denominator expressions for the supply-demand coordination index i, which are related to the constraint function. They are all defined in Affine functions on; These are the lower and upper bounds of the decision variable, respectively.

[0042] The original nonlinear fractional programming problem is transformed into an equivalent mathematical programming problem with a linear objective function and special bilinear structure constraints using an equivalent transformation. Based on the mathematical characteristics of the equivalent problem, a suitable relaxation programming problem is constructed, and the calculation is as follows:

[0043] ;

[0044] In the formula, , These are the lower bound feature and upper bound feature of the objective function, respectively;

[0045] Then introduce Auxiliary variables We obtain an initial outer space relative to the original problem's variable space:

[0046] ;

[0047] In the formula, Let be the initial outer space, and let be the initial feasible domain set of the auxiliary variables; for 3D real space as the initial outer space The carrying space, dimensions, and number of auxiliary variables.

[0048] Assuming Z is the number of indicators with positive coefficients in the objective function, then , The following equivalent problem is obtained from the original problem:

[0049]

[0050] In the formula, (EP D ) represents the equivalent problem function; Let this be the objective function in the equivalence problem; These are constraints in the equivalence problem;

[0051] Assumption For any subregion generated during the partitioning of the outer space, ,when Time definition:

[0052] ;

[0053] In the formula, The constraint function related to the auxiliary variable z in the constructed relaxed linear programming; , Decision variables The adjustment parameters, Positive and negative determination Value rules; , , For auxiliary parameters; and Auxiliary variables In the current molecular region The upper and lower bounds in the range;

[0054] Obtaining the equivalence problem Based on outer space Relaxed linear programming problem :

[0055] ;

[0056] Specifically, based on the relaxed linear programming problem, the outer space branch and bound algorithm is used for solving it, including:

[0057] S1, Initialize the number of iterations Set convergence error Subdivision ratio coefficient Calculate the initial vertex parameter values ​​in outer space. Then, solve the initial relaxation linear programming problem to obtain its optimal solution. and its optimal value ;make: , , , , ;

[0058] if The algorithm stops at this point. These are the questions Find the global optimal solution and optimal value; otherwise, let , , Activity Node Set ;

[0059] S2, For any molecular region Assuming It is the upper bound of the best known optimal value, if ,but No issues included Find the optimal solution, and then delete the sub-region. The corresponding child nodes will still represent the shrunk region as ;

[0060] S3, will the area Divided into two new sub-regions and Delete region and will new molecular regions and Add to the active subset; let: ,definition ,in Let be a constant parameter, and define the interval Divided into two sub-intervals and ;

[0061] S4, to In the respective regions Solve the relaxed linear programming subproblem to obtain the optimal solution. and optimal value Then let Through calculation To determine whether to update the upper bound; if , then delete The corresponding node, if the active node set The algorithm stops; the best known feasible solution is... and its corresponding target value The global optimal solution and optimal value of the original problem are obtained; otherwise, calculation is performed. Update the Nether;

[0062] S5, if The algorithm terminated. and These are the global optimal solution and the global optimal value of the original problem, respectively; otherwise, let Then jump to S2.

[0063] The present invention has the following advantages:

[0064] This invention innovatively transforms maximizing coordination capability into minimizing coordination deviation and solving it in a unified manner with economic cost by constructing a multi-dimensional supply and demand coordination capability evaluation system and a dual-objective optimization planning model. Combined with a linearized global optimization algorithm, it achieves the comprehensive optimization of the economy, security and clean and low-carbon performance of various resource planning in provincial power grids under the high proportion of renewable energy access. Its method system is highly efficient and can be extended to the coordination optimization of energy systems at different levels. Attached Figure Description

[0065] Figure 1 This is a schematic diagram of the comprehensive empowerment process of the present invention;

[0066] Figure 2 This is a schematic diagram of the outer space branch and bound algorithm of the present invention;

[0067] Figure 3 This is a schematic diagram of the net load sampling curve;

[0068] Figure 4 A schematic diagram showing the weight calculation results for each method;

[0069] Figure 5 A diagram illustrating the increased capacity for each device in various scenarios;

[0070] Figure 6 This is a diagram illustrating the calculation results of the indicators in various scenarios;

[0071] Figure 7 This is a schematic diagram of the simulation operation of scenario 1-100 days;

[0072] Figure 8 This is a schematic diagram of the simulation operation of scenario 2-100 days;

[0073] Figure 9 This is a schematic diagram of a typical day's operation under peak load in Scenario 1;

[0074] Figure 10 This is a schematic diagram of a typical day's operation under minimum load in Scenario 1;

[0075] Figure 11This is a schematic diagram of a typical day's operation under peak load in Scenario 2;

[0076] Figure 12 This is a schematic diagram of a typical daily operation simulation under minimum load in scenario 2. Detailed Implementation

[0077] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only for explaining the invention and are not intended to limit the invention; that is, the described embodiments are merely some embodiments of the invention, and not all embodiments. The components of the embodiments of the invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0078] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0079] It should be noted that relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0080] The present invention will be further described below with reference to the accompanying drawings, but the scope of protection of the present invention is not limited to the following description.

[0081] like Figures 1 to 12 As shown, a source-grid-load-storage resource collaborative planning method for optimizing supply and demand coordination capabilities is characterized by the following: the method includes,

[0082] A collaborative capability evaluation index system is constructed to quantitatively assess the collaborative capability of supply and demand. The collaborative capability evaluation index system includes the proportion of new energy installed capacity, the proportion of new energy power generation, the adequacy of installed capacity, the adequacy of reserve capacity, the power utilization rate, the new energy utilization rate, the cost ratio, the proportion of adjustable power installed capacity, the average adjustable capacity of adjustable power, the adequacy of adjustable power capacity, the proportion of imported power, the annual average load satisfaction rate, the energy storage configuration ratio, and the average utilization rate of energy storage.

[0083] Percentage of new energy installed capacity:

[0084] ;

[0085] In the formula, The proportion of new energy installed capacity; , These refer to the existing installed capacity of wind and solar power, respectively. , The newly installed capacity for wind and solar power respectively; This is the sum of the installed power supply capacity of all systems.

[0086] Percentage of electricity generated from renewable energy sources:

[0087] ;

[0088] In the formula, The proportion of electricity generated from new energy sources; For wind power output during period t; Provide photovoltaic power output for period t; This is the sum of the power generation of all power supply devices during time period t; The amount of electricity generated for which electricity is purchased;

[0089] Sufficient installation capacity:

[0090] ;

[0091] In the formula, To ensure sufficient capacity for installation; The load power at each time point;

[0092] Reserve capacity adequacy:

[0093] ;

[0094] In the formula, Sufficient reserve capacity; The sum of power output from sources other than new energy sources at all times;

[0095] Power utilization rate:

[0096] ;

[0097] In the formula, For power utilization; , Predict power output values ​​for wind and solar power at various times; The duration of the statistical period; This represents the actual power generation of a conventional power source. This represents the actual power generation from new energy sources.

[0098] New energy utilization rate:

[0099] ;

[0100] In the formula, To improve the utilization rate of new energy sources;

[0101] Cost percentage:

[0102] ;

[0103] In the formula, Cost percentage; For investment costs; Equivalent cost of existing installations;

[0104] Adjustable power supply installed capacity percentage:

[0105] ;

[0106] In the formula, The percentage of adjustable power supply installed capacity;

[0107] Average adjustable capacity of adjustable power supply:

[0108] ;

[0109] In the formula, The average adjustable capacity of the adjustable power supply; Adjustable capacity for traditional power supplies;

[0110] Adjustable power supply capacity adequacy:

[0111] ;

[0112] In the formula, To ensure sufficient adjustable power supply capacity;

[0113] Percentage of incoming electricity:

[0114] ;

[0115] In the formula, The percentage of electricity generated from external sources;

[0116] Annual average load fulfillment rate:

[0117] ;

[0118] In the formula, This represents the annual average load satisfaction rate. The load shedding amount at each time point;

[0119] Energy storage configuration ratio:

[0120] ;

[0121] In the formula, The energy storage configuration ratio; , These refer to the initial and investment power capacities of electric energy storage, respectively. This refers to the initial and investment power capacity of pumped storage;

[0122] Average utilization rate of energy storage:

[0123] ;

[0124] In the formula, This represents the average utilization rate of energy storage. , , , These are, respectively, energy storage discharge power, pumped hydro storage discharge power, energy storage charging power, and pumped hydro storage charging power;

[0125] This paper employs a combined weighting method, utilizing the CRITIC objective weighting method and the G1 subjective weighting method, to assign weights to indicators. The CRITIC objective weighting method extracts objective weights from precise objective information, while the G1 subjective weighting method ranks indicators based on decision-makers' experience to obtain subjective weights. This synergy ensures that the weights both accurately reflect actual data characteristics and professional judgment. A linear weighting method is used to calculate the combined weights based on the objective and subjective weights, using the following formula:

[0126] ;

[0127] In the formula, For subjective weight vectors, ;w k For objective weight vectors, ; These are the combination coefficients; This is an unnormalized combined weight vector; represents the standardized combined weights; N is the length of the indicator matrix, and n is the indicator number.

[0128] A comprehensive optimization planning model that balances economic efficiency and supply-demand synergy is constructed based on a collaborative capability evaluation index system; the comprehensive objective function is to minimize the total life cycle cost and maximize the supply-demand synergy.

[0129] The objective function of the comprehensive optimization planning model is:

[0130] ;

[0131] In the formula, For investment costs; Operating costs; For carbon emission costs; C buy For electricity purchase cost; C cut Cost of power curtailment; This is the upper limit of the cost. A quantitative value for supply and demand coordination capability; This is the comprehensive weight matrix of the indicators, which consists of the weights of 14 indicators obtained by the combined weighting method.

[0132] ; This is a vector of index values; , The weights are respectively based on economic efficiency and the ability to coordinate supply and demand, among which It is a negative number.

[0133] Investment costs:

[0134] ;

[0135] In the formula, , , , , These are the investment costs for the capacity of gas-fired power, coal-fired power, wind power, photovoltaic power, and energy storage systems, respectively. The adjusted capital recovery factor; G, C, W, P, These are the installed capacities of gas-fired power, coal-fired power, wind power, solar power, and energy storage systems, respectively.

[0136] Operating costs:

[0137] ;

[0138] In the formula, , , , These are the operating cost coefficients for gas-fired power, other thermal power, coal-fired power, and nuclear power units, respectively. , , These are the adjustment cost coefficients for gas-fired power, other thermal power, and coal-fired power ramp-up, respectively. , , , These are the unit base power generation outputs for gas-fired power, coal-fired power, thermal power, and nuclear power, respectively. , , These refer to the regulated power generation output of gas-fired power, coal-fired power, and thermal power, respectively.

[0139] Electricity purchase cost:

[0140] ;

[0141] In the formula, This is the unit cost coefficient for purchased electricity;

[0142] Cost of curtailment:

[0143] ;

[0144] In the formula, The unit penalty cost coefficient for abandonment of renewable energy; The unit cost coefficient for load shedding;

[0145] Carbon emission costs:

[0146] ;

[0147] In the formula, , , These are the carbon emission cost coefficients for gas-fired power, coal-fired power, and other types of thermal power, respectively.

[0148] The constraints of the comprehensive optimization programming model include:

[0149] System equilibrium constraints:

[0150] ;

[0151] In the formula, P buy,t The amount of electricity generated from purchased electricity; where P is the power generated. buy,t The amount of electricity generated for which electricity is purchased; , , , These are, respectively, energy storage discharge power, pumped hydro storage discharge power, energy storage charging power, and pumped hydro storage charging power; This represents the sum of the power generation of all power supply devices during time period t.

[0152] Shear load constraint:

[0153] ;

[0154] In the formula, The load shedding amount at each time point; The load power at each time point; This represents the maximum load shedding rate at each time point; This represents the overall maximum load shedding rate.

[0155] Conventional power supply dynamic constraints:

[0156] ;

[0157] In the formula, A collection of conventional power supply units; Let t be the online capacity of a certain power unit; , To enable or disable capacity; This refers to the rated capacity of a power unit. To provide power to a certain type of power unit; This represents the minimum technical output rate of a certain power unit. and These are the minimum and maximum ramp rates for this power unit;

[0158] New energy constraints:

[0159] ;

[0160] In the formula, For wind power output during period t; Provide photovoltaic power output for period t; , The predicted power output values ​​for wind and solar power during time period t are respectively.

[0161] Constraints of energy storage systems:

[0162] ;

[0163] In the formula, It combines electrical energy storage and pumped hydro storage; , These are the maximum power and capacity of the energy storage, respectively. Indicates the lower limit of energy storage capacity; , These represent the states of charge of a certain energy storage system at time t and the previous time, respectively. For time step; , The charging power at time t and time t-1 are respectively. , These represent the discharge power at time t and the previous time, respectively. Let be the state variables of the operating mode of a certain energy storage system during time period t; The overall energy conversion efficiency of a certain energy storage system.

[0164] The optimal solution is obtained by solving the comprehensive optimization planning model.

[0165] The planning problem considering the integrated capacity of supply and demand coordination is a nonlinear programming problem, and it is also an affine fractional sum problem—for example, the proportion of adjustable power sources, the utilization rate of new energy sources, and the utilization rate of energy storage all require fractional functions of decision variables. The nonlinear part of the integrated optimization planning model is transformed into a linear problem through a linearization method for solution, as calculated below:

[0166] ;

[0167] In the formula, Let x be any real number; x is the decision variable; For the decision variable space; function , These are the numerator and denominator expressions for the supply-demand coordination index i, which are related to the constraint function. They are all defined in Affine functions on; These are the lower and upper bounds of the decision variables, respectively. Solving this type of problem directly falls under fractional programming, which is non-convex, and conventional gradient-based or heuristic algorithms struggle to guarantee global convergence.

[0168] The original problem is transformed into an equivalent mathematical programming problem with a linear objective function and special bilinear structure constraints using an equivalent transformation. Based on the mathematical characteristics of the equivalent problem, a suitable relaxation programming problem is constructed, and the calculation is as follows:

[0169] ;

[0170] In the formula, , These are the lower bound feature and upper bound feature of the objective function, respectively;

[0171] Then introduce Auxiliary variables We obtain an initial outer space relative to the original problem's variable space:

[0172] ;

[0173] In the formula, Let be the initial outer space, and let be the initial feasible domain set of the auxiliary variables; for 3D real space as the initial outer space The carrying space, dimensions, and number of auxiliary variables.

[0174] Assuming Z is the number of indicators with positive coefficients in the objective function, then , The following equivalent problem is obtained from the original problem:

[0175]

[0176] In the formula, (EP D ) represents the equivalent problem function; Let this be the objective function in the equivalence problem; These are constraints in the equivalence problem;

[0177] According to (EP) D The inner space (decision variable space) and outer space (auxiliary variable space) The dimension can be determined by choosing a space with a smaller dimension for partitioning, and then establishing a relaxed linear programming problem with equivalent problems in that space. Typically, in practical problems, the inner space dimension is larger than the outer space dimension; therefore, the outer space is chosen here. For ease of description, let's divide it into parts, assuming... For the subregions generated during the partitioning of the outer space, a method for estimating the linear sub-variable of the nonlinear function in the construction problem (EP) can be obtained through a series of basic operations and transformations. This method is applicable to any... ,when Time definition:

[0178] ;

[0179] In the formula, The constraint function related to the auxiliary variable z in the constructed relaxed linear programming; , Decision variables The adjustment parameters, Positive and negative determination Value rules; , , For auxiliary parameters, where, This indicates that the original constraint is related to... Directly related auxiliary parameters; It is a scalar constant that reflects the offset of the constraint intercept or constant in the original problem. Auxiliary parameters for auxiliary variables; and Auxiliary variables In the current molecular region The upper and lower bounds in the range;

[0180] Obtaining the equivalence problem based on outer space Relaxed linear programming problem:

[0181] ;

[0182] Every feasible solution to a relaxed linear programming problem is also an equivalent problem. A feasible solution, whose optimal value does not exceed the equivalent problem. The corresponding objective value, therefore, the relaxed linear programming problem is an equivalent problem. It provides an effective lower bound for the optimal value, and relaxes the linear programming problem as the partitioning process is performed an infinite number of times. This will infinitely approach the equivalence problem .

[0183] In the source-grid-load-storage resource collaborative planning for optimizing supply and demand coordination capabilities, the optimization model is characterized by high dimensionality and nonlinearity due to the involvement of system modeling with multiple devices, time periods, and indicators. Through equivalent transformations and variable substitutions, the original fractional programming problem is transformed into an equivalent problem with a linear objective function and bilinear constraints. Although the objective function has been linearized, bilinear terms of type z⋅x still exist in the constraints, causing the problem to remain nonconvex after transformation. In this case, directly using a general linear programming solver cannot handle the bilinear constraints; if conventional nonlinear programming methods are used, convergence to a local optimum may occur. Based on the relaxed linear programming problem, an outer space branch-and-bound algorithm is used for solving it. This algorithm systematically partitions the outer space and constructs a relaxed linear programming problem in each sub-region, gradually approximating the optimal solution of the original problem. Its mathematical mechanism guarantees convergence to the global optimum within a finite number of steps. Simultaneously, through the constructed linear relaxation problem, a lower bound of the optimal value for the current sub-region can be obtained in each iteration. Combined with upper bound updates and region pruning, regions that do not contain the optimal solution are quickly eliminated, improving solution efficiency.

[0184] S1. Initialization; Initialize the number of iterations. Set convergence error Subdivision ratio coefficient Calculate the initial vertex parameter values ​​in outer space. Then, solve the initial relaxation linear programming problem to obtain its optimal solution. and its optimal value ;make: , , , , ;

[0185] if The algorithm then stops. These are the questions Find the global optimal solution and optimal value; otherwise, let , , Activity Node Set ;

[0186] S2, Region Reduction; for any partitioned region Assuming It is the upper bound of the best known optimal value, if ,but No issues included Find the optimal solution, and then delete the sub-region. The corresponding child nodes will still represent the shrunk region as ;

[0187] S3, Branch; This will divide the region. Divided into two new sub-regions and Delete region and will new molecular regions and Add to the active subset; let: ,definition ,in Let be a constant parameter, and define the interval Divided into two sub-intervals and ;

[0188] S4, Bounding; for In the respective regions Solve the relaxed linear programming subproblem to obtain the optimal solution. and optimal value Then let Through calculation To determine whether to update the upper bound; if , then delete The corresponding node, if the active node set The algorithm stops; the best known feasible solution is... and its corresponding target value The global optimal solution and optimal value of the original problem are obtained; otherwise, calculation is performed. Update the Nether;

[0189] S5, Optimality test; if The algorithm terminated. and These are the global optimal solution and the global optimal value of the original problem, respectively; otherwise, let Then jump to S2.

[0190] Using the actual power grid data of a certain province in 2024 as boundary data, the installed capacity of various resources is shown in Table 1 below. In addition, the external power grid access capacity is 39,300 MW, with a planning target of 2030, the predicted maximum load in 2030 is 190,000 MW. Total thermal power, wind power, photovoltaic power, and energy storage are used as the proposed investment and construction equipment in the planning model. The installed capacity boundaries of each equipment are shown in Table 1.

[0191] Table 1 Installation Boundaries of Each Equipment

[0192] To ensure the representativeness and comprehensiveness of typical scenarios across time, this paper employs a three-dimensional stratified sampling method: month-season-cluster. Specifically, firstly, an improved K-means clustering method is used to divide the annual daily load curve into several typical clusters. Then, the frequency of occurrence of each cluster is statistically analyzed by month and season, serving as the basis for weighting. Finally, typical scenarios for a corresponding number of days are extracted from each cluster according to the weighted proportions, constructing a 100-day scenario set. The sampling results are as follows: Figure 3 As shown.

[0193] The weight calculation results for each method are as follows: Figure 4 The calculation results show that different methods characterize the weights of the indicators differently. The combined weighting method optimizes the weight distribution while retaining subjective and objective information through weighted harmonization, thus supporting the balanced consideration of the multi-dimensional objectives proposed in the article.

[0194] The weighting logic of the dual objective function of this invention continues the CRITIC-G1 combination weighting logic of the index layer, resulting in an economic weight a of 0.382 and a supply-demand coordination capability weight b of 0.618.

[0195] To more clearly compare the impact of supply and demand coordination capabilities on the optimization planning of provincial power grid resources, the text sets up two different scenarios for comparative verification analysis.

[0196] Scenario 1: Planning scenario considering only economic factors;

[0197] Scenario 2: Planning scenarios that consider both economic efficiency and supply-demand coordination.

[0198] To solve this optimization problem, this paper uses the MATLAB and YALMIP programming environments to establish a mathematical model and calls the Gurobi software package for solution. The solution times for scenarios 1 and 2 are 806s and 1140s, respectively.

[0199] A comparison of planning results under different scenarios reveals a significant divergence in system resource optimization strategies under different planning objectives. This divergence is directly reflected in the layout of new installed capacity for various power sources and energy storage systems. The planned new installed capacity under each scenario is as follows: Figure 5 As shown.

[0200] from Figure 5 As can be seen, Scenario 2 represents a low-carbon transformation of its power structure compared to Scenario 1. Its newly added thermal power capacity is 51% lower than in Scenario 1, while the newly added renewable energy capacity is 108% higher. Scenario 2, influenced by its "clean and low-carbon" indicators, strengthens the system's low-carbon attributes, aligning with the transformation direction of a high-proportion renewable energy power system. Simultaneously, Scenario 2's newly added energy storage capacity is 4.92 times that of Scenario 1. By significantly increasing energy storage capacity, Scenario 2 enhances the system's buffering capacity against uncertainties in wind and solar power output and reduces reliance on thermal power for peak shaving. To further quantify the impact of these resource allocation differences on the system's overall performance, an in-depth analysis is conducted combining cost and indicators. The planning costs for each scenario are shown in Table 2, and the indicator calculation results are as follows: Figure 6 As shown.

[0201] Table 2. Planning cost calculation results for each scenario (unit: RMB 100 million)

[0202] From a cost perspective, the total cost of Scenario 2 is higher than that of Scenario 1, mainly due to a significant increase in investment costs, reflecting the initial investment requirements for large-scale deployment of high-proportion new energy and energy storage. However, at the same time, the operating costs and carbon emission costs of Scenario 2 are reduced by 25.2% and 39.7% respectively compared to Scenario 1, confirming the economic advantages of low-carbon transformation in long-term operation.

[0203] In terms of performance indicators, the installed capacity adequacy in Scenario 2 is 34.8% higher than in Scenario 1, indicating a significant improvement in the matching degree between the total installed capacity and the maximum load, providing basic capacity support for high-load scenarios. The significantly improved reserve capacity adequacy in Scenario 2 greatly enhances the system's buffering capacity against fluctuations in wind and solar power output and sudden load increases, reducing the risk of supply-demand imbalance. To reduce reliance on local thermal power, Scenario 2 optimizes the local supply-demand balance by introducing external power, resulting in a higher proportion of imported electricity in Scenario 2 compared to Scenario 1.

[0204] Operational characteristics analysis

[0205] To further verify the supply and demand coordination effectiveness of different planning schemes in actual operation, this paper selects a typical 100-day scenario for operational simulation analysis. The operational simulations for each scenario are as follows: Figure 7 and Figure 8 As shown.

[0206] Comparing the operational simulations of Scenario 1 and Scenario 2, in Scenario 1, thermal power, due to its high new capacity and singular economic focus, exhibits a characteristic of consistently high, rigidly supported output: thermal power output exceeds renewable energy output for extended periods, only slightly reducing load during off-peak hours, with its annual operation mode approaching that of a "baseload power source." In Scenario 2, the output curves of renewable energy and thermal power alternate and complement each other. During periods of high renewable energy generation, thermal power plants significantly reduce their load to allow for renewable energy absorption; during periods of low renewable energy generation, thermal power plants quickly increase output to fill the power gap. This model aligns with the planning stage's focus on ample safety and clean, low-carbon development.

[0207] In Scenario 1, energy storage exhibits characteristics of "low frequency, low power, and passive supplementation," only charging a small amount of surplus power from new energy sources or supplementing short-term load gaps during very few periods, without deeply participating in system scheduling. In Scenario 2, however, energy storage exhibits characteristics of "high frequency, high power, and active collaboration," actively charging a large amount during periods of high new energy generation and actively discharging a large amount during periods of low new energy generation or peak load, becoming a core flexible resource of the system and effectively mitigating the uncertainty of new energy output and the peak-valley difference in load.

[0208] Analysis of operating characteristics under extreme conditions

[0209] Under extreme operating conditions, each power source exhibits different timing-coordinated response mechanisms. To accurately capture the operational differences of the system under extreme load conditions, a comparative analysis was conducted using the lowest load day and the highest load day within a 100-day simulation period. The typical daily operation simulations for scenarios 1 and 2 under extreme loads are as follows: Figures 9-12 As shown.

[0210] On typical days with peak load, in Scenario 1, thermal power maintains a relatively stable high output level, slightly substituting for thermal power output during periods of limited renewable energy generation. Due to limitations in deployment scale and planning guidance, energy storage only makes minor adjustments during periods when renewable energy generation peaks overlap with secondary peak loads. In Scenario 2, thermal power and renewable energy complement each other. During periods of high renewable energy generation, thermal power plants significantly reduce their load to allow for renewable energy absorption. When peak load coincides with periods of low wind and solar power activity, thermal power plants quickly respond to fill the gap, while energy storage deeply participates in power balancing. This, combined with on-demand response from purchased electricity, forms a multi-source linkage system to ensure power supply resilience.

[0211] On typical days with the lowest load, the dispatch strategies for the two scenarios are basically the same. Thermal power plants also operate to their minimum technical output boundary; energy storage plants charge and discharge more frequently, absorbing surplus power from new energy sources through high-frequency energy regulation, forming a dynamic power interaction with thermal power and new energy sources to jointly adapt to the operational needs of the off-peak load scenario.

[0212] The above description is merely a preferred embodiment of the present invention and does not constitute any limitation on the present invention. Any person skilled in the art can make many possible variations and modifications to the technical solution of the present invention, or modify it into equivalent embodiments, without departing from the scope of the present invention. Therefore, any modifications, equivalent changes, and alterations made to the above embodiments based on the technology of the present invention without departing from the scope of the present invention are within the protection scope of the present invention.

Claims

1. A resource coordination planning method for source-grid-load-storage resources oriented towards optimizing supply-demand coordination capabilities, characterized by: The method includes, Construct a collaborative capability evaluation index system to quantitatively assess supply and demand collaborative capabilities; Based on the collaborative capability evaluation index system, a comprehensive optimization planning model is constructed that takes into account both economic efficiency and supply-demand collaborative capability. Its comprehensive objective function is to minimize the total life cycle cost and maximize the supply-demand collaborative capability. The optimal solution is obtained by solving the comprehensive optimization planning model.

2. The source-grid-load-storage resource collaborative planning method for optimizing supply and demand coordination capabilities according to claim 1, characterized in that: The collaborative capability evaluation index system includes the following indicators: proportion of new energy installed capacity, proportion of new energy power generation, adequacy of installed capacity, adequacy of reserve capacity, power utilization rate, new energy utilization rate, cost ratio, proportion of adjustable power installed capacity, average adjustable capacity of adjustable power, adequacy of adjustable power capacity, proportion of imported power, average annual load satisfaction rate, energy storage configuration ratio, and average energy storage utilization rate.

3. The source-grid-load-storage resource collaborative planning method for optimizing supply and demand coordination capabilities according to claim 1, characterized in that: It also includes a comprehensive weighting method that uses the CRITIC objective weighting method and the G1 subjective weighting method to assign weights to indicators. The CRITIC objective weighting method is used to obtain objective weights by accurately mining objective information, while the G1 subjective weighting method is used to rank the importance of indicators based on the decision-maker's experience to obtain subjective weights.

4. The source-grid-load-storage resource collaborative planning method for optimizing supply and demand coordination capabilities according to claim 3, characterized in that: Using linear weighting, a combined weight is obtained based on objective and subjective weights, calculated as follows: ； In the formula, For subjective weight vectors, ;w k For objective weight vectors, ; These are the combination coefficients; This is an unnormalized combined weight vector; is the standardized combined weight; N is the length of the indicator matrix; n is the indicator number.

5. The source-grid-load-storage resource collaborative planning method for optimizing supply and demand coordination capabilities according to claim 1, characterized in that: The comprehensive objective function is: ； In the formula, For investment costs; Operating costs; For carbon emission costs; C buy For electricity purchase cost; C cut Cost of power curtailment; This is the upper limit of the cost. A quantitative value for supply and demand coordination capability; This is the comprehensive weight matrix of the indicators; This is a vector of index values; , The weights are respectively based on economic efficiency and the ability to coordinate supply and demand.

6. The source-grid-load-storage resource collaborative planning method for optimizing supply and demand coordination capabilities according to claim 5, characterized in that: The constraints of the comprehensive optimization programming model include: System equilibrium constraints: ； In the formula, P buy,t The amount of electricity generated for which electricity is purchased; , , , These are, respectively, energy storage discharge power, pumped hydro storage discharge power, energy storage charging power, and pumped hydro storage charging power; This is the sum of the power generation of all power supply devices during time period t; The load shedding amount at each time point; The load power at each time point; Shear load constraint: ； In the formula; This represents the maximum load shedding rate at each time point; This represents the overall maximum load shedding rate. The duration of the statistical period; Conventional power supply dynamic constraints: ； In the formula, A collection of conventional power units; Let t be the online capacity of a certain power unit; , To enable or disable capacity; This refers to the rated capacity of a power unit. To provide power to a certain type of power unit; This represents the minimum technical output rate of a certain power unit. and These are the minimum and maximum ramp rates for this power unit; New energy constraints: ； In the formula, For wind power output during period t; Provide photovoltaic power output for period t; , The predicted power output values ​​for wind and solar power in time period t are respectively; Constraints of energy storage systems: ； In the formula, It combines electrical energy storage and pumped hydro storage; , These are the maximum power and capacity of the energy storage, respectively. Indicates the lower limit of energy storage capacity; , These represent the states of charge of a certain energy storage system at time t and the previous time, respectively. For time step; , These are the charging powers at time t and time t-1, respectively. , These represent the discharge power at time t and the previous time, respectively. Let be the state variables of the operating mode of a certain energy storage system during time period t; The overall energy conversion efficiency of a certain energy storage system.

7. The source-grid-load-storage resource collaborative planning method for optimizing supply and demand coordination capabilities according to claim 5, characterized in that: The nonlinear part of the comprehensive optimization programming model is transformed into a linear problem using a linearization method, and the calculation is as follows: ； In the formula, Let x be any real number; x is the decision variable; For the decision variable space; function , These are the numerator and denominator expressions for the supply-demand coordination index i, which are related to the constraint function. They are all defined in Affine functions on; These are the lower and upper limits of the decision variable, respectively; The original nonlinear fractional programming problem is transformed into an equivalent mathematical programming problem with a linear objective function and special bilinear structure constraints using an equivalent transformation. Based on the mathematical characteristics of the equivalent problem, a suitable relaxation programming problem is constructed, and the calculation is as follows: ； In the formula, , These are the lower limit and upper limit features of the i-th indicator, respectively; Then introduce Auxiliary variables We obtain an initial outer space relative to the original problem's variable space: ； In the formula, Let be the initial outer space, and let be the initial feasible domain set of the auxiliary variables; for The dimensional real space, which serves as the initial outer space The carrying space; Assuming Z is the number of indicators with positive coefficients in the objective function, then , ; We obtain the following equivalent problem to the original problem: ； In the formula, (EP D ) represents the equivalent problem function; Let this be the objective function in the equivalence problem; These are constraints in the equivalence problem; Assumption For any subregion generated during the partitioning of the outer space, ,when Time definition: ； In the formula, The constraint function related to the auxiliary variable z in the constructed relaxed linear programming; , Decision variables The adjustment parameters, Positive and negative determination Value rules; , , For auxiliary parameters; and Auxiliary variables In the current molecular region The upper and lower bounds in the range; Obtaining the equivalence problem Based on outer space Relaxed linear programming problem : 。 8. The source-grid-load-storage resource collaborative planning method for optimizing supply and demand coordination capabilities according to claim 7, characterized in that: Based on the relaxed linear programming problem, the outer space branch and bound algorithm is used to solve it, including the following steps: S1, Initialize the number of iterations Set convergence error Subdivision ratio coefficient Calculate the initial vertex parameter values ​​in outer space. Then, solve the initial relaxation linear programming problem to obtain its optimal solution. and its optimal value ;make: , , , , ; if The algorithm then stops. These are the questions Find the global optimal solution and optimal value; otherwise, let , , Activity Node Set ; S2, For any molecular region Assuming It is the upper bound of the best known optimal value, if ,but Does not include equivalence issues Find the optimal solution, and then delete the sub-region. The corresponding child nodes will still represent the shrunk region as ; S3, will the area Divided into two new sub-regions and Delete region and will new molecular regions and Add to the active subset; let: ,definition ,in Let be a constant parameter, and define the interval Divided into two sub-intervals and ; S4, to In the respective regions Solve the relaxed linear programming subproblem to obtain the optimal solution. and optimal value Then let Through calculation To determine whether to update the upper bound; if , then delete The corresponding node, if the active node set The algorithm stops; the best known feasible solution is... and its corresponding target value The global optimal solution and optimal value of the original problem are obtained; otherwise, calculation is performed. Update the Nether; S5, if The algorithm terminated. and These are the global optimal solution and the global optimal value of the original problem, respectively; otherwise, let Then jump to S2.

Images